Show case memos.web.app

[fabriziogiordano]

This is not an issue. I just wanted to thank you.
I create a memos web app that transcribe your memos.
Nothing fancy but with you code I can make it run on a small shared server.

Url:
https://memos.web.app/

Git:
https://github.com/fabriziogiordano/memos

[fabriziogiordano]

Closing since this was not really an issue.
But I hope @ggerganov will check it out :)

[ggerganov]

Awesome!
I converted the issue into a discussion in case other people also find it interesting.

[jafri]

What model size is this using?

[fabriziogiordano]

Thank you @ggerganov your work is just amazing.

@jafri it is currently using the tiny model. Here is the source.

This is pretty much a proof of concept. The bottle neck here is the server I use to do the transcoding. It is the cheapest server available on Vultr (ps I am not sponsored at all).

I used the tiny simply for the transcoding speed. Base was ok but a little slow for the web app. I will try to deploy a stronger server (with at least 2 CPU and more memory).

The current CPU specifications:

>lscpu
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   40 bits physical, 48 bits virtual
CPU(s):                          1
On-line CPU(s) list:             0
Thread(s) per core:              1
Core(s) per socket:              1
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       AuthenticAMD
CPU family:                      23
Model:                           49
Model name:                      AMD EPYC-Rome Processor
Stepping:                        0
CPU MHz:                         1996.251
BogoMIPS:                        3992.50